PRIORITISATION OF TARGET AREAS FOR FOREST RESTORATION Final Report Produced by WCMC for WWF International May 2000 YAN 5U PRIORITISATION OF TARGET AREAS FOR FOREST RESTORATION Contents ABACK GNOUMG: Serrecttee cease teers cct oboe neconven ase stemesuacuces cUsrentecerstaursesctonetesrsacterccentsss tecenseszsacerseesusvercesrenarz= c= 1 2. Objectives SiillargetaudienCe sis. severe c.-s-csscesecvevseassner AT AD FIM ILIONS os oes socesnes cvs azssceseesseseescccesvsemereees Bi, CICILY CTL) ceca ceseceee nee pose bose op pccaocoencccbanodsocnbocdae oe s06c5000¢ GOODE ROJO BEBD4s Hope uDoGodEd0=0Cc os BocHooaoRscGoECIH0—ea 2 6. Consultation........... a SR EE Mees ceca 6 7. AMAlySiS ........cccececeeceececeescsceseseesensnscsssscscseseresesesnensenssenscnseseateucnasucsacensassesssecessesssserenseceasesensenseseatenscaees 6 £4, Cray p el UST S ce scosceeeccoeosaecaenace er os00c0000osancconoconsedeoocEeoecededaeeaDodsca Hee bee SHEEN Baa BEE BE DeCERqa on aceneoCCoALAcaCcGGECaEHd 7 QMREGOMINIEMGALOMS rertrrrcre crc eres ee ane e ee nnomateer acne sstontevona stone canes tecmcnccscse= 7 [PRETENCE tesco co cece eobanee80sc 68) pend ao RESEaTiEod suo decoSgaSsogbe50 Joocooaa co EEdaebaoccaSsedancchoce osshlacesondecosauogaec=ccogqageGnacccdasananc 8 1. Background Recent estimates place current global forest cover at 50% of its original extent, with a large proportion of this loss having occurred within the past 50 years. Despite conservation efforts, many of the remaining forest ecosystems have been seriously degraded and fragmented, resulting in environmentally, economically and aesthetically impoverished landscapes. Loss of forest cover not only limits the scope of biodiversity and genetic conservation but also diminishes the prospect of many rural and coastal populations throughout the developing world attaining a decent and secure livelihood. In many areas, faltering clean water supplies, catastrophic floods and landslides, declining fish stocks and unreliable local weather patterns can be traced back to impaired forest ecosystem functions. WWF and IUCN are working to develop a co-ordinated programme of work aimed at mobilising public and private sector funds into socially and ecologically appropriate forest restoration initiatives. The programme will aim to provide guidance to governments and industry on good forest restoration practice and on definition of priority areas for such activities. 2. Objectives The objectives of this study are to: » identify candidate criteria that can be used for prioritising socially and ecologically appropriate forest restoration at a regional level; = define broad areas which can provide a focus for a regional restoration programme. The Mediterranean region was used as a case study. However, the methodology was developed to be applicable, with modification, in any of the other five WWF target areas for restoration (the Lower Mekong, India, East Africa, North West Europe and Central America). The study does not set out to define actual restoration sites. Such sites requires extensive local knowledge and cannot be identified using coarse (0.1-1.0 km?) resolution imagery. 3. Target audience The target audience for this study comprises: s WWF International and WWF regional offices; « other organisations requiring a national or regional decision-making tool for prioritising restoration initiatives. The study is not designed to assist decision-makers working on forest restoration projects at the local level. 4. Definitions In this study, we use definitions set out in the WWF report: Global Forest Restoration: a review. Reforestation is defined as the managed reestablishment of forests on lands that once contained forests. Reforestation activities produce different types of forest which may be listed in order of increasing ecological integrity: deforested land plantation forests reclaimed forests transitional forests rehabilitated forests restored forests native forests Forest restoration therefore involves the establishment of forests containing native species with a high ecological integrity. Restoration is defined as the deliberate alteration of ecological patterns and processes for the purpose of recreating some presumed set of natural, pre- disturbance ecosystem conditions. Restored forests are therefore similar in structure, function and composition to historic forests. 5. Candidate criteria There are many ecological and social criteria which could be used to identify priority areas for forest restoration. The selection of criteria depends principally on issues concerning the nature of data and is constrained by: = availability of data (whether restricted by ownership, cost or coverage); = resolution of data (whether suitable for global, regional, national or local analyses); = quality of data (whether reliable and/or relevant to the task in hand) = analysis required (whether or not processing can be carried out within the budgeted time frame). Twelve candidate criteria were considered on the basis of data heid at WCMC. With reference to the constraints set out above, five were selected (see Table 1). Others were considered to have potential, but could not be used given the time constraints of the study. The need to identify global data sets greatly limited the range of criteria for selection. Selection of criteria must also reflect forest restoration objectives, in particular: = the desired benefits from forest restoration (whether for biodiversity conservation, recreation, watershed protection and/or production); = the preferred sites for forest restoration (whether on bare sites, within woodland mosaics, around forest boundaries or within degraded forests). Objectives were not specified by WWF for the purposes of the study. It was therefore necessary to make certain justifiable assumptions to limit the scope of analysis (see Table 2). io punos6 Buidojs Ajyua6 40 yey UO Sealy “uoN|OSe1 WH | & UM (OEOdOLD) POM au} JO japowW UOHeAge ey6IP ey} WO PeAJap uaeq aAeYy pjnoo dew ajbue edojs v ‘sessejo Ajlsuap jo Jaquunu Aue oyu! paljissejoeJ aq ued eJep ay} ‘sylun Usemjeq uOHejOdiaju! Aq pauleygo SeM UOHepeld (GEGL 72 J 418/G01) seljuNoo UIUJIM SHUN sArey}SIUIWpe JO} Byep WO peonpoid uaaq sey dew Ajisuep uoNjejndod ay) Ayisuap uolyeindod MO| JO Sed *seaie Ajsesojoibe pue uoeyeban jesnyeu JO seove yueoyluBIs yjIM esnynouBe Aq paidnooo Ajjediouud pue| ‘seinysed sapnjou! eunyjnoube @AISUB}X8 JO UOHIUYEP By, “SUJaWed UOHeAHIND xajdwoo ‘sdoJO yUOUeWEd UYyIM pej}eloosse sdoso yenuue ‘seaoi6 anljo ‘saysnq pue see} }iny ‘spseAOUIA ‘splay ao ‘pue] ajqese payeb1u! Anueuewied ‘pue ajqese peyebusi-uou soepnjoul aay pesn oinynoube sAlsuejul Jo uOHIUYEp BY) *selpoq Jayem pue sjey jepeju ‘sues ‘shoq ‘sayssew ‘mous jenjadied pue siaioe|6 ‘pue| usueg = seale pajeyaban Ajasueds pue sease jung ‘spue|yjeoy ‘sioow ‘spuejsses6 jenjen = aunyjnowHBe sAlsua}xy = aunyjnowBe sAisua}u| = :(uole10}S91 JO} ApIIGeVINS Jo Japs0 Hulpusosep ul) Sassejo p OjU! payissejoes sian sdew esn pue] jeulblio ay} UO pasn sesse|D pp OY, (WI 0Gz :uoNnjoses) sdew asn pue] 3NINOO ey} Woy peAUep useq aAey Sdew asn pue| paljl|dWis peyseiojun Ajyuasno ede YoIyM Ajisuayul asn pur] Mo| jo seauy “spaysieyeM J0/pue uojepeibep [IOS UO UONEWIOJU! Y}IM pasn 4! jNJdjay aq Aew ejyep 3u} ‘IOAQMOH ‘(Sedojs deajs epiy Aew ulei9} yBnod ul! {Wy} Jenusajod JO Base ue) 9SJEOD 00} UO!}N|OSEJ BUL seH - | paydao0yv “BVeEPp SOUS pue SNIYOO jog a}epowwos9e Pinos yoiyum ABojopoujew uoneoyissejoas e dojanap 0} ajqeyiene Owl} JUBIOYJNSU! SEM BJO] ‘@aHeIBACD Jeqo|6 apinoid 0} SON] Woy eyep YIM pe}uswajddns aq pjnoo eyep FNINOO ‘JEAQMOH *(SeljJUNOD UeadoiNy awWoS pue ued YON ‘Usa}Sez s|pply JO} | jenua}od eyep ou) ajajdwoou! s! aHesaAo0o0 aul sey - | paydscoyv - | _paydesoy uol}9afas 10} UOSeaY | UOISIDEG "$90J} }INJ 10/pue Sdos9 JepuN pue| [Je Sepnjou! pue| ajqese ‘puejqniys jou ng pue|ysng pue }@91Uu} Sepnjoul pue|poom :mopeew pue ainjsed sapnjou! asay pesn puejssesb Jo uoH!Uyep OY “‘pue] UeGIN JO pue}eM ‘pue| Useg ‘Wasep-lwWas ‘Weseq «= {pue|POOM JNOU WM pueqniys JO/pue pue| ajqese JO/Pue PUR|SSEID = ‘PUE|POOM UM pue] ajqese JO/pue pue|SSel5 = :(UONeIO}SOJ JO} AIQe}INS Jo JOpsO Bulpusoseap ul) sesse|o ¢ OU! payissejoas asam Sdew JaA09 puke] jeulbi0 au} UO pasn sesse|o COZ SY “SO!ESOW pue|POOM JO pue| papoom Ajjemed yybyybiy 0} (uoNNjosel zy | @ YIM sebew! ey\jayes pey}esdsajul) sdew J9A09 pue|] SOYA Woy peuluuajap useq sey seaJe jsel0j Bulpunowns esn pueq “pa}sas0jas 8q PINOD YDIUM seale jo jua}xe aiqissod wnwixew 94} sjuaseides JaA09 }seJ0} jeUIBUG “uOHe}eHen jenua}od jo sdew wo. paauep si pue ofe syeaX QOOL punose paysixa YOIYM JBAOD yseJ0} JeIDe|6-jsod au} Se peuep si jseio} jeuIBUC ‘sdew JaAoopue_| (WI OOO} :uOoNN|Ose’) pue WSJ ‘15S (wW 0SZ :UONNJOSA1) ANINOD Woy peAusp useq sey uoibel UB@ULIOE}IPA/\) SU} JO} JOAOD SOO} JUBLIND SINDM 32 Play B}ep UO UOHJeULOJU] peyseuojun Ajjuaino ese YOIUM pue|poom Bululeyuo0o seal pe}sauojun Ajyuasind ase YOIUM sease }saJ0} jeulBUC VIMALIYD J@AQ] |euOIHaJ e Je UO!}EJO}SO1 }SOJOJ JO} Seae Ajluold Huluyap Jo} Eley ayepipued :} B1qeL seae JUSWIYD}JED JOALY eBueyo O}EWI|D WO) JEOIU} Japun jou seay uolj998}01d pasiubo09e1 Japun Apeasje seaiy sedfjoiq ANIHOO se peuyep sjuiod punoie seaiy seaie psig jueyoduy| “sisAjeue pue yoseasal JOuyNY auinbad pjnom yoy ‘(uoHepesbap jlos Ajejnonied) seaie yueuyoyeo JENPIAIpul UO UOHEWIOJU! payeloosse | jenuajod ‘suoiBas UBaMjaq |12}9P Ul SALEA S}JUBWYD}ED JAPJO JAMO] UO UO!JEWUJOJU] }NOUYM |Nydjay you ae eyep ay sey | ‘(ééé 90un0s) paonpoid uaeq aney savepuNog sjusWYo}ed JOAN PUe SJOAW UO eJeEp jeqo|g “Ajeuoibe. -qns parjdde aq 0} asueod 00} Ajqissod *(P9}D9]9S aq 0} 819M UOL9}UO SIU} J! UOI}E10}S91 SI SjJapowW UOH}e}aH9A Jo uO!Njosal }SOJ0} JO} |]qe}yINSUN 9q PjnomM UeEaUeE}IPayY] BU} Jo JsOW yeu} Jeaedde pinom }j ‘suONoIpeid ay] “awed atu} pejeHpng ou} ulyym | jenuajod | Huysixa usAId) “uoNN|Osel saiHep | e YM sjapow asuodses uoNe}eHen payeloosse pue }NO paseo aq jou pjnoo sisAjeue sy sey | ebueyo ayew|d snouen ul payljuap! uaeq aAey aHueyo ayeuljo Aq payoaye Ajasanas ySow Sealy “aiqeyiene osje si sease pajoa}oud payeubisap Ajjeuoneusayul ewes} awl} payoHpng ay} ulyyIM | jenuajod UO UOHEWIOJU] *(|A-|) S8U0H9}e9 NON] 0} Buipsoooe payissejo aie sease payoajoid payeuBbisap }NO pales aq jou pjnoo sisAjeue ay sey AlJEUOHEN “OWOM Aq paddew pue paj}99]/09 useq aAey sease pa}oa}oud UO e}ep jeqo|D swa}sksooe pajeloosse uO UONEWOJU! JNOUY!M [Nydjay you aie eyep “UO!}EWOJU! ay, (Saljunod ueadoiny awos pue pojyeloosse ynoUjIM BWOdD nq ‘suo!}e90] JUIOd avey sadAjoiq ay) ,UoNe/si6a; Aq pajoajoid UBD YWON ‘SCJ a/ppip\ 40} eyep ou AIJEUUIOY S| Bae SIY] JBYJOYM Jo SsajpseBas ‘UuOHeAIaSUOD aiNjeU JO, BOUeDYIUBIS UeadoinZ-ued Jo ade 91984}) aja|dwooul s! aHesan0d au | payelay | yun seaBojo9a ue SuOy yoIym Apog 13}eM e JO puke] JO Bale, Ue Se pauyap si edAjJOIG FNINOD V “ApN}S Siu} JO} SWI SU} OU} UIUJIM peyesauUab aq jOU pjnod YoIUM ‘Slua}SASO09 Po}elN0SSe UO UO!}EWOJU! ynouym [nydjay you aie eyep ay *(BOLY UON JO} OWOM Aq pley aie “UONEWUOJU! P9}BINO0SSe }NOY}IM BWOD jnq ‘suO!}ed0] JuIOd e}ep Ou) a}a|dwooul si aesaAo0d ay] | payosfay | uaAih aie segue psig juUeHOdU| ‘}|euOHeWE}U] O4!|P4Ig Aq peyUap! Useq eAeY seaie pig JUeHOdUW] "saujunoo ueedoing -UOU U! payjijUap! UBaq JOU SAeY S}Ods}oY }SaJ0} pue payIwo useq eAey Aew sjods}oy uMoUy -||AM SSO] ‘JOABMOH “UO!E1O}SJ JSO1OJ JO} SNDOJ jnJasn e se avuas AewW Sease YONS ‘AjISJBAIPOIG }SOJ0J JO SoujUBO JUeLOdUII Se Spadxa }S9J0} |eD0] Aq payljuap! 19M S}ods}oy eseu| ‘pappe usEeq a@aey s}ods}oy }se10} UeadoIN>F 4M\M ‘JEP BY} SAOIdWI O| "UOHEIO}SAI }S9IO} 10} Seaie AyWOUd Bunoajas 10} siseq |njasn e aplAoid ||}s AjIsJOAIp Ue] JO SeujUBd BU} ‘“JIOABMOH “(SWi9a}SASO99 9ue}JUOW UO SISeydia ay} BdUaY) Saldeds }saJ0} JO AjISIBAIP UE} JOUJeJ |e19USBH Ul Saloads jue|d jo Ayisuanip ayeoipul paddew sease ay] "}U9}SISUODU! BIE $9J}UBD JO UOI}D|9S JO} EL9}9 uy} pue Hulyoe| Ss! uoIbas UBSULA}IPA/\ BU} JO} [!e}ap yeU} SUBSW yD9fOId OU} JO aunyeu jeqo|H oul - | paydecoy (p66) NONI pue 4MM) 399!01d 4MM 1egoI6 e Aq peuyep useq aAey Ajissanip jue|d Jo SasjUuaD *(S]@S E}EP S910} JO UONNjOSa! Ou} 0} Bulpuodsai09 ‘SOUJUNOD JOY}O JO} LY | ‘SayUNOD UeSdOINy JO} W OGZ :wWNwiUIW) paweA aq UeD U;WPIM JayjNq - | paydacoy | ayy “Wy, JO U}PIM JayNg e YYWM dew JOAOD }S9JO} }UBIND BUY} WO. payesoudH useq sey dew Vv AISJBAIpOlq Ul YO Sealy $}SO10} Bunsixe 0} Ayiuixoid @SO}O Ul SPAY jOA9Q| |euoIGad e ye UO!}eJ0}S9J 1S9J0} 10} Sease AyJOd Buluyap 40} eL9a}IJ0 aJepIpUeD :(panuuOd) 1 ajqeL UO!}E1O}SO1 JSO1O} JO} “goueyodwi Sanuoud asojasay} aie AjIsaAIpoIq Ayissanipoiq yBiy yo sease UO pasnodoj aq Pinoys sHOYa UO!}e10}SO1 OS ‘UOIFEAI@SUOD AjISIOAIPOIG 0} Ps}INs }S9JO} Ul YOIs Seaue SAjISIOAIPOIG sso] Bulag ysjluMm ‘AJaANDaJe-}SOD SOW S}YaLIaq BSeu} Ja\jO UD s}se10j pobeuewW JEYyjJO ‘JEAeMOH “JEquI} @AJQSUOD O} SI UO!}E1O}SOI Ayss9lpoiq JO UONONpodd (payul|) UBAS JO Spayssayem JO UONDa}ONd se Yons ‘s}yaUeq JayJO aAeY ABW S}SE10} PalO}SOY S910} Jo esodind jediouid ay) ul YO Sealy “uO}e10}SO1 40} sanuoud ase Aepunog }sal0} “spue]| J9U}O UO S}Ule;}SUOD BulUUE|d pue ‘AjIIIGe}INS O}1S 0} ANP s}sa10} BuIsIxe punoue paesnooy | 94} WO WY | > Pue S}ses0) HuNSIxe | s}Ses0} 0} AjiWIxoid aq 0} Ajay!| ase SUOYS UONEeJO}SaI ‘saloads sANeU JO SJIOAIASOI 0} AjIWIXOJd JO BdUeLCdW! 84} O} UONIPpe ul} 0} usoe[pe Ajajyeipowwi sea asojo ul Sealy “y *“SEOJE POPNIOXA JO JU9}X9 OU} }IWI| 0} pasies uVEq sey Ayisuap uonjeindod pjoysesy} ay} JeASMOH *SedJNOSAI }S910} [290] UO BDUapUad~ap jo yOe| Aq peusyeam S$! UONE|a09 Siu} ‘UOIGas UBAUBE}IPEY) 4} JO SELjUNOD padojersp au} U| “Ajisuap UOHe|Ndod Ybiy jo sease Ul UOHEJO}SO1 |NJSsedons 40} saniiqissod ay} Bulonpes ‘(poomeiy pue syonpod }sas0} Jequul}-UOU JO} “6'e) sa0Jnosal jeinyeu uO ainsseid pue Ajisuap uoe|ndod usemjeq UOHe|asJ09 BsojO e S| aJau} SaljUNOD AUeW U| *UO!}E10}S9J SOO} JO} B]qeyINsuN aue (uoibes uesueEe}Ipey| OU} 10} Wy jad suosiad 008<) Ajisuap uojeyndod Ayisuap uoyjejndod yBiy jo seaiy MO| JO SPAY "€ “UONEJO}JSOJ }SOIO} JO} B]Qe}INS eue puejsses6 10/pue pue] ajqese | paysasojun Ajjuasino Ul pue|POOM Jo sayojyed HululeyuoD | ase YOIUM pue|POOM $}S910} Buljsixo apis}no sealy Bululeju09 sealy ‘pa10}$8J @q |II}S Pinod ynq ‘asn puke] aAlsua}u! BJOW JUasesdal SPUE|POOM Jenpisel OU YIM pue|sses6 pue pueR| ajqeiy ‘payuawbey AjyBiy si UoYNquysIp 119U} J! UBAS ‘puNoy |INS ole Saloeds pue|POOM aAljeU aBYyM Seale U! BAHOaYe Pue a|qISeE} }SOLU aq 0} Aj9y!| S! UONEJO}SOI }SO104 “sesodind Jauj}o 10} pebeuew $}S9JO} PU S}SO10} jeINJeU ‘Way} UsaMjaq YsiNBul}sIP 0} BjGe 9q 0} UO!IPUOD }SeJO} UO yep Us!OyNSU! SI e184) 3Ng ‘UOI}EJO}SE JO} SalOd eq Aew s}saso} papeiBap Auew ‘joes ul ‘aye}S |eunjeu JY} 0} PeAel 0} ajqe aq 0} salads anlyeu Jo SuaquNU yUaIOWINS UleEJUOD (J9A0D AdOUeD %O]}-< YIM S}SeJOj |euOHISUes} pue Jano0o Adoued %OE< U}IM BSOU} Se peuljep) seae ysol0} AuewW ‘JUaWeHeUeW anl}oe sopuN Apeaije JOU BJEYyA\ “UOI}E1O}SOJ }SO1OJ JO} Jenuajod avey pa}sesojap Mou a1e nq }se10} payoddns aou0 YyoIyM seaJe |e ‘uoie10}S9J 10} Ayold yBiy & jou ae seale }Se10} JUDLIND pa}salojun Ajyuasind ase YOIUM seale }sai0j jeulbO suonduuinsse 10} uojeoynsne suonduinsse jes VIMALINO Jag] |euOIBa & Je UO!}J0}S9J }S9JO} JO} Sease AyJOUd Bulujap JO} eLa}9 ayepipued payoeIag :7 a1qeL 6. Consultation WWF experts were asked to comment on the criteria selected for identification of restoration priority areas. A discussion document and map transparencies (which could be used in combination to show how each criterion affected the extent of priority areas) were provided. Feedback from the reviewers has been extremely useful in improving assumptions and selection and analysis of candidate criteria. A number of points were made by the reviewers, which fell into two categories: those relating to the strategic objectives of forest restoration and those relating to the criteria that could be used for defining priority areas. The need for identification of clear objectives for forest restoration before selection of criteria takes place was noted by most reviewers. Criteria used to identify areas suitable for projects with biodiversity-focused objectives will be very different from those for projects with timber production, protection of water supplies or prevention of desertification as their main objectives. To complicate matters, all restoration activities will be affected by the policies and incentives of national governments. The objectives of restoration projects throughout the Mediterranean region are therefore likely to vary considerably and it is unrealistic to expect generic criteria to provide accurate mapping at a national level. Nevertheless, most reviewers believed that indicative mapping of priority areas at a regional level was a useful exercise. A range of additional criteria were thought to merit consideration. These included: « forest and/or ecosystem type legal status of forest (gazetted or non-gazetted) land tenure historical land use patterns soil degradation vegetation condition river basins socio-economic factors (e.g. rural economy) Whilst these criteria could not be used in this study, their importance is recognised. If the methodology applied here were to be used to define priority areas at a national (rather than regional) scale, it would be extremely useful to include some or all of these criteria. However, in this study we have deliberately excluded data which are specific to the Mediterranean so that comparable analyses can be carried out for other regions. 7. Analysis A Geographical Information Systems (GIS) analysis was used to define priority areas according to the criteria listed. Each of the five criteria defined in Table 2 was applied consecutively to reduce the area with potential for restoration. In effect, areas which were not considered to have restoration potential were subtracted from the map at each step (see Table 3). The order in which the criteria were applied / output images were produced is a (subjective) reflection of their importance in identification of priority areas. The (untested) assumption is that the criteria applied at the beginning of the analysis are less controversial than those applied at the end. Areas of plant and biodiversity are considered to be relevant to every output image. However, the centres of plant diversity are too broad and the forest hotspots too limited (point locations only) to figure in the analysis. The biodiversity hotspots map has therefore been used as an overlay image in conjunction with each map. Copies of the maps (numbered / lettered as in Table 3) may be found at the end of the report. Table 3: Flow chart showing steps in GIS analysis Input map images Output map images (showing areas Overlay map images with forest restoration potential) A. Original and > 1. Land area originally forested but E. Biodiversity hotspots current forest cover currently without forest cover S v 2. Land area identified in Map 1 where B. Current land cover > | current landcover includes some € | E. Biodiversity hotspots woodland mosaic v 3. Land area identified in Map 2 where C. Population density > | population density <800 inhabitants € | E. Biodiversity hotspots per km? v D. Proximity to forests > 4. Land area identified in Map 3 that is € E. Biodiversity hotspots <1 km away from current forest cover v 5. Composite map prioritising areas for e E. Biodiversity hotspots forest restoration 8. Conclusions The composite map showing priority areas for forest restoration for the Mediterranean is a first attempt to produce such a regional scale prioritisation. As the selection criteria evolve, the candidate criteria identified in this study will require modification. However, the most urgent next step is the comparison of predicted priority areas on the map with the location projects on the ground; this will determine whether the criteria selected here reflect current policy on project establishment and, if so, where additional projects may be needed. Selection of criteria for prioritisation depends on the objectives for forest restoration. Crucially, should restoration focus on areas most threatened by human activities (soil erosion, forest exploitation, pollution), or is it more appropriate where there is less pressure on forest ecosystems? There are arguments for both courses of action. The former may yield more conservation benefits, the latter may be more cost-effective. As WWF and other conservation organisations develop regional forest restoration policies, the regional priority-setting tool developed here can be improved. Selection criteria must then reflect how forest restoration will be implemented. Will restoration efforts build on existing forest, link fragmented woodlands or establish new forests from bare sites? A number of assumptions have been made in this study to limit the scope of the analysis; in fact, it may be necessary to develop alternative priority maps for each possible course of action. 9. Recommendations To improve the applicability of the regional approach described in this report, we make the following recommendations: « Identification and mapping of current forest restoration activities, to test the extent to which project location conforms to the rationale used in this study; « Assessment of objectives of local forest restoration projects, to validate assumptions made and improve the selection criteria (do these reflect objectives set by international bodies working at a regional scale?); « Reproduction of maps after WWF have agreed regional priorities; = Repetition of the mapping process for other WWF regions; « Use of additional criteria excluded in this study through lack of time for analysis. One issue that has not been addressed in this study is the application of the GIS approach used here at a sub-national level. For many parts of the world, spatially-referenced data at this scale are unavailable; however, the Mediterranean region has been the subject of more research and untapped data sources do exist. Using sources of information other than those discussed here, it may be possible to develop a tool for planning of local restoration activities’. To this end, we make the following recommendations: * Development of the land-cover layer to show land-use and land-use intensity more clearly; in the case of Europe, agricultural practice is changing and much former military, industrial and agricultural land has a potential for forest restoration. =» Improvement of the accuracy and precision of existing land cover data with updated AVHRR data; » Use of detailed information on location and status of protected areas for restoration planning and prioritisation; « Use of spatially-referenced information on forest condition (e.g. as a result of harvesting or fire damage), local conservation priorities and forest management practices. References Tobler, W. Deichmann, U., Gottsegen, J. and Maloy, K. 1995 The Global Demography Project. National Center for Geographic Information and Analysis, University of California, Santa Barbara. WWE and IUCN 1994 Centres of Plant Diversity. A guide and strategy for their conservation. 3 volumes. IUCN Publications Unit, Cambridge, UK. "WCNC is currently working with Dr Sandra Luque from the University of Cambridge to develop a landscape level GIS methodology to define forest restoration priority areas for two countries in the Mediterranean region. SNOIABY “OD Aq paydwoD pas upiuayssd he =o101apy Ae Bea”? ae — | Sion d4dNn yse10) |BUIBUQ(s yse0) ueUND ages Y3AAO9 LSAYNOS LNSYYND GNV TWNISTHO 000z ABW ‘ SWOM dann © SNOIABY “D Aq pajdwod re oWOM ddNN Jan00 JSa404 JnoUgUA AjUELINS jng peyseuoy AyeulBuo puey] || ) ) TWILNSLOd NOILVHOLS3AY LSAYOS HLIM SVAYNV 0002 ABW ' SWOM dann © 9 9 9 SNOIIABY “SD Aq payduioD ot = Pn pag uvjaunsdajipapy pas upiuayssdy Be S21IDIApy uva2Q a14uD]1P OJWOM ddNn = y pug] uequn pue ]BSOW PUB|POOM jHOUTM Solesow puB|pooM TRIM yw rT) puenjem ‘ueueq eae ornaaTieeat Beno }sel0) 0800 > Eee] puejqniys roe pug} cae ieee pu} ahs Tonle Moiese ( i YAAODGNV1T LNAYYND entific 0002 Aew ‘ WOM daNnN © Ea \ ) { } SNOIIABY “SD Aq poyduiod pas upiuoy 4 (siujpres 4 pag yon] Sy a up220 o14uD/IP S| : sen 3 Me, ees Sent ed \ ‘ Le C ‘ ys wees 1 hora meen fp % Ge BTN on 1 orandaliaza». ., oWwom dann o® DIBSOW pUe|POOM eWOS SepnjoU! JeACOpUR] JUEUIND e1eYM ‘| dew ul payuep! Bere pue|] a CG TWILNSLOd NOILVYOLSSY LSAYOS HLIM SVAYNV ee ee = a een eet Sour ; one a in r +) Ou 000z Aew ‘ SWOM dann © Ww SNayiAey “OD Aq peydwod Qaud44A9aq Tasers ine > ays oos-to> «= oo-t0z RS soz- 10 crt ALISN3G NOILV1NdOd ‘wy ‘bs sed squeyiqeyu c . alk uy o | - : Ses cle ree oie [Fr SNoWAeY “D Aq pay dwod 0002 ABW " OWOM daNN © wy ‘bs sed syugyiqeyu! Og > Ajisuep uoneindod eseym ‘Zdew u! peyyuep! Bele puey pers WOM dann 9 cy WILN3LOd NOILVYOLS3AY LSSYOS HLIM SVANV 0002 ABW ' SNOM daNn © 5 wag en “AS unasay “308 SS Wom dann (O) (es) JOAOD }S@J0j JUEUIND JO WY} UIYYM PUey asset @| SLS3YOS OL ALINIXOUd ees SF Se ; = mee, _ = aa eee: Caen: fer 3 “Toe 9 2 Ore ee ee oe 5 ST eae Is ree rs ie 3 i = a S~ 5 : SNOIIABY “D Aq pajidwod 000z Aew ' SWOM dann © pag yo01g ees N Bes Spc a ; ; rPrapy\ JIM dINQ sen09 18010) Jueuno Woy Aeme uN) > Ss} eu) ‘¢ dew ui peynuepl eve puey] | TWILN3LOd NOILVHOLSSY LSSYOS HLIM SVSNV yy ee | a = as ae ee sau SNOWAeY °O Aq paidwog SNOWIAGY “D Aq pondwid aS Fae Kn upiuayasdy uy : te Ae 5 ~ PCO AP | YAN (4M Aq paynuepr se) sjods}0} jse10. SJOAID JUB|d JO SE.QUED ests [enuejod ysemoy Tenuejod yseyB}y =a wruk 7 tet Nes @TOVeTOH YTI2AavIGOla (AWW va betlinsbi 2s) atogatoH tes107 © — tierevib inlqo eetneD es@ Aonla diab h psa atime lyk nns30 SNOWIABY °D Aq paydwod ' 1 1 L 4 000z Aew ‘ OWOM dann © pag ubdiaudddajZipapy pas upiuoy Dag yovi|g requejod ysemo7 | ))/ NINN renucjod seus G TWILNSLOd NOILVYOLSSY LSSYOS HLIM SVAYV AO M3SIANSAO oWom dann